Bintrafusp alfa, a first-in-class bifunctional fusion protein composed of the extracellular domain of TGF-βRII (a TGF-β “trap”) fused to a human immunoglobulin G1 monoclonal antibody blocking programmed death-ligand 1 (PD-L1), has exhibited clinical activity in a phase 1 expansion cohort of patients with PD-L1–high advanced NSCLC. This adaptive phase 3 trial (NCT03631706) compared the efficacy and safety of bintrafusp alfa versus pembrolizumab as first-line treatment in patients with PD-L1–high advanced NSCLC. Primary end points were progression-free survival according to Response Evaluation Criteria in Solid Tumors version 1.1 per independent review committee and overall survival. Patients (N = 304) were randomized one-to-one to receive either bintrafusp alfa or pembrolizumab (n = 152 each). The median follow-up was 14.3 months (95% confidence interval CI: 13.1–16.0 mo) for bintrafusp alfa and 14.5 months (95% CI: 13.1–15.9 mo) for pembrolizumab. Progression-free survival by independent review committee was not significantly different between bintrafusp alfa and pembrolizumab arms (median = 7.0 mo 95% CI: 4.2 mo–not reached (NR) versus 11.1 mo 95% CI: 8.1 mo–NR; hazard ratio = 1.232 95% CI: 0.885–1.714). The median OS was 21.1 months (95% CI: 21.1 mo–NR) for bintrafusp alfa and 22.1 months (95% CI: 20.4 mo–NR) for pembrolizumab (hazard ratio = 1.201 95% CI: 0.796–1.811). Treatment-related adverse events were higher with bintrafusp alfa versus pembrolizumab; grade 3/4 treatment-related adverse events occurred in 42.4% versus 13.2% of patients, respectively. The study was discontinued at an interim analysis as it was unlikely to meet the primary end point. First-line treatment with bintrafusp alfa did not exhibit superior efficacy compared with pembrolizumab in patients with PD-L1–high, advanced NSCLC.